ML Systems Research Engineer, RL / Inference / Agent Systems

Build reinforcement learning, inference, and evaluation infrastructure that enables agentic engineering workflows. You’ll design scalable systems for long-horizon, high-latency rewards, staged reward evaluation, and uncertainty-aware metrics. Partner with AI research scientists to develop reward and post-training methods, while implementing distributed experimentation, job orchestration, caching, experiment tracking, and reproducible evaluation pipelines for production engineering teams.